Which theoretical framework posits language learning emerges from patterns of activation in interconnected networks resembling neural networks?

Which theoretical framework posits language learning emerges from patterns of activation in interconnected networks resembling neural networks?

• A. Connectionist/Parallel Distributed Processing
• B. Universal Grammar
• C. Transformational Grammar
• D. LAD

Answer: (A)

Connectionist/Parallel Distributed Processing models propose that language learning emerges from patterns of activation across interconnected networks. In these frameworks, linguistic knowledge isn’t stored as explicit, symbolic rules but as distributed representations—the strengths of numerous connections among units that adjust as a learner is exposed to language input. Through gradual updates, the network learns to recognize and produce language by capturing statistical regularities in the input, allowing generalization to new sentences without relying on pre-specified rules. This mirrors how neural networks operate, with knowledge arising from patterns of activation across many parts of the system. Universal Grammar and Transformational Grammar, in contrast, emphasize innate grammatical principles and rule-based transformations, while the Language Acquisition Device posits an inherent mechanism for language learning. These frameworks focus on internal mental representations and predefined rules rather than learning through distributed patterns of activation, so the described mechanism aligns best with the connectionist/PDP view.

Connectionist/Parallel Distributed Processing models propose that language learning emerges from patterns of activation across interconnected networks. In these frameworks, linguistic knowledge isn’t stored as explicit, symbolic rules but as distributed representations—the strengths of numerous connections among units that adjust as a learner is exposed to language input. Through gradual updates, the network learns to recognize and produce language by capturing statistical regularities in the input, allowing generalization to new sentences without relying on pre-specified rules. This mirrors how neural networks operate, with knowledge arising from patterns of activation across many parts of the system.

Universal Grammar and Transformational Grammar, in contrast, emphasize innate grammatical principles and rule-based transformations, while the Language Acquisition Device posits an inherent mechanism for language learning. These frameworks focus on internal mental representations and predefined rules rather than learning through distributed patterns of activation, so the described mechanism aligns best with the connectionist/PDP view.